Proposition of a Typology of Tunisian Consumers Based on their Ecologically Conscious Behavior

International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 360 Proposition of a Typology of Tunisian Consumers Based on their Ecologically Conscious Behavior Faouzia Mida 1 Department of Management, ESSECT, Tunis, Tunisia Imed Zaiem Department of Management, FSEGN, Nabeul, Tunisia Abstract The environmental movement led companies to consider ecology as a strategic decisional criterion. Concerns have started to be integrated in the study and practice of management and marketing. The object of this research is to propose a typology of Tunisian consumers in terms of their ecologically conscious behavior. More specifically, this work describes the identified groups based on socio-demographic variables (such as gender, age, income, socioprofessional category and religiosity) and on the psychographic variable perceived consumer effectiveness. To this end, we conducted a questionnaire survey next to 200 consumers. The empirical results show the existence of four consumer profiles which are the convinced, aware, mitigated and resistant which led us to produce some useful recommendations for marketing managers wishing to be positioned on the ecological axis. Keywords: ecologically conscious consumer behavior, perceived consumer effectiveness, socio-demographic variables, typology. Cite this article: Mida, F., & Zaiem, I. (2015). Proposition of a Typology of Tunisian Consumers Based on their Ecologically Conscious Behavior. International Journal of Management, Accounting and Economics, 2(5), 360-380. Introduction During the last decades, ecology has become a social priority in which various actors share responsibilities (producers, consumers, citizens, government). Environmentalist movements drive the companies to consider ecology as a strategic decisional criterion. Such concerns started to be involved in the study and practice of 1 Corresponding author’s email: [email protected] International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 361 management and marketing. Within the scope of marketing research, this movement is called « Ecological Marketing» or « green marketing». Green marketing covers all the areas that could generate and facilitate any type of exchange aiming at satisfying human desires and needs in such a way that the latter’s satisfaction takes place with a minimum negative impact on the natural environment (Stanton and Futrel, 1987). Thus the term « green » is associated to the environment or to ecology and refers to problems linked to the air, water and earth. The term « green » is mainly used to describe consumers with ecological sensitivity (Hackett and al., 1991). At the level of the literature, concerns on ecology were treated through various approaches. Some studies explored the consumption schemes with ecological responsibility such as knowledge of ecological products (Cronwell and Schwepker, 1995; Balderjahn, 1988) as well as the purchasing motivations of an ecological product (Kréziak, 1995). Others were interested in the conceptualization and operationalization of concerns for the environment (Giannelloni, 1998). Other studies were interested in the description of the profiles of individuals who are concerned with ecology through the examination of socio-demographic and psychographic characteristics of the respondents (Pickett et al., 1993; Kréziak and Valette-Florence, 1997). Our research is inscribed within the framework of the last approach. In fact, several studies tried to describe the green consumer (Scott and Willits, 1994; Martin and Simintiras, 1995; Shrum and al., 1995; Awad, 2011); however, to this day, there is no commonly accepted profile hence the need for the present research which tries to propose a typology of consumers in terms of their ecologically conscious behaviors (ECCB: Ecologically Conscious Consumer Behavior). The description of the identified groups will be done in addition to the sociodemographic variables (such as gender, age, income, socio-professional category or SPC and religiosity) and the psychographic variable « Perceived Consumer Effectiveness ». This will help propose useful recommendations to marketing managers hoping to position themselves in the ecological axis. Theoretical Framework Environmental Concern (EC) There have been various definitions of environmental concern proposed in the literature. These definitions depend on the complex and unstable nature of the term (Chan and Lau, 2004). Maloney and Ward (1973) conceptualize environmental concern as being the totality of knowledge about ecology, the degree of affection towards ecology and the level of verbal commitment « ecological intention » and of concrete commitment « ecological behavior» to ecological problems. Crosby and al. (1981) define environmental concern as an extremely protective attitude towards the environment. To Chan and Lau (2004), EC is synonymous to attitude towards the environment. International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 362 Studies on concern for the environment dealt with several levels of analyses. Some studies focused on the evaluation of specific behavior such as recycling, green consumption, and energy conservation. (Arbuthnot, 1977; Pickett and al., 1993). Others dealt with the evaluation of general behavior (Antil, 1984; Roberts, 1996a and b). Therefore, there is no ground to compare the conducted studies. In fact, comparing two studies can only be done if they are about the same level of analysis. The analysis of the literature shows the links between environmental concern and some ecological behaviors (Roberts, 1996; Chan and Lau, 2000). For instance, Bamberg (2003) shows that the degree of environmental concern has a direct and strong influence on behavior in terms of recycling and energy saving, choice to purchase environmentally friendly products or even in the way people travel. Bang and al. (2000) stipulate that consumers who are more concerned with environmental issues are more ready to pay higher prices for renewable energy than those who are less concerned. Kim and Choi (2005) highlight that it is more probable that people who are more concerned about environmental issues buy environmentally friendly products than those who have less concerns in this respect. The literature review on environmental concern indicates that researchers have not managed to explain the social bases of an ecological behavior whence the necessity to establish the profile of ecological consumers. Profile of the ecological consumer Ecological marketing literature tried to identify the characteristics of the profiles of individuals who are concerned with ecology. The first attempts date back to the 1960s when Berkowitz and Lutterman (1968) studied the profile of socially responsible consumers. In fact, a socially responsible consumer is « a consumer who takes into account the public consequences of his/her private consumption and who tries to use his/her purchasing power to introduce change in society » (Webster, 1975). Starting from the 1990s (called « earth decade »), interest in ecology was considerable on the academic level. Research then focused on the description of the consumer starting from his/her ecological behavior. The works of Roberts (1996a) highlighted the concept of « ECCB: Ecologically Conscious Consumer Behavior ». Roberts indicated that the consumer is considered as environmentally concerned if he/she buys products and services that have a positive impact or that have less negative impacts on the environment. In order to better describe green consumers, several studies tried to define their characteristics. The results of the studies were contradictory in the way some found significant results and others found insignificant results (Do Paço and Raposo, 2009). The literature review shows that the socio-demographic variables help frame the profile of the green consumer. In fact, these variables offer an easy tool for announcers to segment the market (Anderson and al., 1974; Samdahl and Robertson, 1989; Roberts, International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 363 1996 a and b; Jain and Kaur, 2006; D’Souza and al., 2007). Among the sociodemographic variables that were object of ecological behavior study we find : age (Van Liere and Dunlap, 1981 ; Pickett and al., 1993), gender (Baldassare and Katz 1992, Brooker, 1976), income (Anderson and al. 1974 ; Antil 1984 ; Van Liere and Dunlap, 1981), education (Webster, 1975 ; Kinnear and al., 1974), place of residence (Zimmer and al., 1994 ; Schwartz and Miller, 1991), household size, density of the population (Antil 1984) and economic and social status (Antil,1984 ; Vining and Ebreo, 1990). In addition to the socio-demographic variables, psychographic variables better explain and predict ecological behavior (Straughan and Roberts, 1999). Among the psychographic variables having been the object of study on ecological behavior, we find: political or ideological orientation (Hine and Gifford, 1991; Roberts, 1996a), perceived consumer effectiveness (PCE) (Fraj and Martinez 2006), alienation (Anderson, Henion and Cox, 1974), center or perimeter of control (Dolich and al., 1981) and socialization (Webster, 1975). Thus, D’Souza and al. (2007) describe the green consumer as a young educated welloff citizen. Schwepker and Cornwell (1991) describe the green consumer as one with a high income and higher education diploma. To Webster (1975), the green consumer is a dominant, tolerant and responsible person. He/she is also a person who is highly persuaded that he/she can reduce pollution problems and takes into consideration the ecological impact of his purchases (Kim and Choi, 2005; Lee and Holden, 1999). This is what we call « perceived consumer effectiveness ». To Arbuthnot (1977), this consumer is rather a person with a liberal political ideology. Anderson, Henion and Cox (1974) define the green consumer as one with a more alienated and cosmopolitan personality, less conservative and dogmatic. However, these few profiles cannot establish a particular ecologically conscious consumer profile and therefore confirm the inexistence of a commonly accepted one (Shrum and al., 1995; Tilikidou and Zotos, 1999). The proposition of an ecologically conscious consumer typology becomes then necessary. This is what we will demonstrate within the scope of our empirical study. Methodology Our study aims to identify a typology of consumers in terms of their ecologically conscious behaviour. To this end, we conducted a questionnaire survey and the data were collected through the administration of questionnaires via face to face interviewer next to 200 consumers. Our sample was selected using the convenience method. However, we have tried to have a structure of age and gender that is similar to the Tunisian population structure. You will find the characteristics of our sample in appendix 2. Regarding the choice of measuring instruments, we have selected Roberts’ scales (1996a). The first evaluates ecologically conscious consumer behaviour and is composed of items reflecting the degree of commitment of the consumer who only buys products that are less harmful to the environment. The author reports a Cronbach alpha of 0.96. International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 364 This scale was also used by Awad (2011) in the context of ecologically conscious consumer segmentation. The second scale allows measuring the perceived consumer efficiency (PCE) which is composed of four statements that reflect consumers’ attitudes to environmental concerns and their estimation that their actions and purchases can make a change (see appendix 1). These various measures were graded from 1 (completely disagree) to 5 (completely agree). The survey’s data were treated using SPSS 18 and AMOS 18 softwares. In fact, we have adopted an exploratory factorial analysis in order to verify the unidimensionality of each of the theoretical constructs and to test their reliability at a later stage. Then, we have determined the results of the confirmatory analysis in the hope of verifying the reliability and validity of the constructs based on Fornell and Larcker’s methodology (1981). Finally, we classified the dynamic multitudes in order to elaborate a typology of consumers in terms of their ecologically conscious behavior. The obtained consumer segments are described by socio-demographic variables (such as age, gender, place of residence, religiosity, PCS and income) and by the psychographic variable « perceived consumer effectiveness ». Analysis and interpretation of results Analysis of ecologically conscious consumer behavior (ECCB) Exploratory and confirmatory analyses present and specify the measurement of ecologically conscious consumer behavior. We have conducted an exploratory analysis (PCA) in order to be reassured on the unidimensionality of the scale’s reliability. In fact, table 1 shows the results of the principal components analysis as well as the internal coherence. This analysis produced 2 axes that explained the ECCB. Consequently, the scale is bidimensional. Axis 1 is called « ecologically conscious purchasing behavior » and axis 2 is called « energy preservation and recycling behavior ». International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 365 Table 1 Principal Component Analysis Items Quality of representation Components after varimax rotation Ecologically conscious purchasing behavior Energy preservation and recycling behavior ECCB3 0,561 - 0,645 ECCB5 0,617 0,779 - ECCB7 0,662 0,814 - ECCB8 0,553 0,743 - ECCB9 0,641 - 0,724 ECCB10 0,696 0,659 - ECCB11 0,678 - 0,820 ECCB12 0,719 0,842 - ECCB13 0,576 - 0,676 ECCB14 0,625 0,766 - ECCB15 0,715 0,803 - ECCB16 0,572 0,542 - ECCB21 0,864 0,890 - Cronbach Alpha 0,907 0,70 K.M.O (Kaiser-Meyer-Olkin) = 0,774 Bartlett’s sphericity test = 0,000 Variance explained in %= 62,134% Table 1 indicates a satisfying KMO value justifying the presence of a factorial solution for both dimensions. Bartlett’s sphericity test is significant to a threshold of 5%. Besides, the quality of items’ representation is good for the items retained except for ECCB1, ECCB2, ECCB4, ECCB6, ECCB17, ECCB18, ECCB19, ECCB20, and ECCB22 that were eliminated. Table 1 equally shows that the Cronbach Alpha coefficients () are reliable at the exploratory level for the dimensions « ecologically conscious purchasing behavior » and « energy preservation and recycling behavior ». Then the confirmatory analysis allows us to verify the reliability and validity of the obtained dimensions (Table 2). Therefore, we have opted for the Fornell and Larcker procedure (1981) in order to calculate the convergent validity and the discriminant validity of the constructs. Table 2 Reliability and Validity of Scale Dimensions Reliability (Rho of Jöreskog) Convergent validity Discriminant validity Ecologically conscious purchasing behavior 0,86 0,76 0,76> 0,21 0,66> 0,21 Energy preservation and recycling behavior 0,92 0,66 International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 366 As shown by table 2, Jöreskog’s Rhô indicated satisfying values exceeding the minimum threshold of 0.7. The Rhô of the convergent validity is superior to the minimum prescribed threshold of 0.5 for each of the obtained dimensions. The discriminant validity was verified as the extracted mean variance is superior to the square of the correlation between the latent variables. We can then conclude that the obtained dimensions are reliable and valid. Table 3 shows that the measuring model of the ECCB presents a good adjustment. Table 3 Adjustment Model Indication Chi-square / ddl GFI AGFI RMR RMSEA NFI CFI Value 1,773 0,911 0,944 0,069 0,074 0,938 0,925 Analysis of the perceived consumer effectiveness (PCE) We conducted an exploratory analysis (PCA) in order to ensure the unidimensionality and the reliability of the scale. Table 4 Principal Component Analysis Items Quality of representation Components EPC1 0,820 0,906 EPC2 0,542 0,646 EPC3 0,544 0,662 EPC4 0,884 0,940 K.M.O (Kaiser-Meyer-Olkin) = 0,658 Bartlett’s sphericity test = 0,000 Variance explained in %= 64% Cronbach Alpha = 0,803 Table 4 indicates a satisfying KMO value justifying the presence of a factorial solution. Bartlett’s sphericity test is significant to a threshold of 5%. The quality of representation is good as all the perceived consumer effectiveness measurement indicators have a value that is superior to the minimum threshold of 0.5. The percentage of the explained variance drives us to retain only one component. Table 4 shows that the Cronbach Alpha coefficient () is reliable at the exploratory level. Using a confirmatory analysis (FCA) validates the measurement model and ensures the reliability and validity based on Fornell and Larcker procedure (1981). Table 5 shows that the measurement model presents a good adjustment by referring to the obtained values of absolute, incremental and parcimonious indices. International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 367 Table 5 Adjustment Model Indication Chi-square / ddl GFI AGFI RMR RMSEA NFI CFI Value 2,914 0,989 0,976 0,029 0,068 0,987 0,991 Table 6 below determines the values of Jöreskog’s Rhô, Rhô of the convergent validity as well as the discriminant validity. These indicators evaluate reliability and validity of the obtained dimensions of the measurement model. Table 6 Reliability and Validity of Scale Scale Reliability (Rho of Jöreskog) Convergent validity Discriminant validity EPC 0,77 0,82 0,82> 0,21 By reading table 6, we conclude that the value of Jöreskog’s Rhô is superior to 0.7 for the measurement model. Equally, the Rhô of the convergent validity is superior to the minimum threshold of 0.5. Finally, we note that the discriminant validity is ensured as the extracted mean variance is superior to the square of the correlation between the latent variables. We can therefore judge that the scale is reliable and valid at the confirmatory level. Identification and specification of consumers’ profiles in terms of their ecologically conscious behavior. We classified the dynamic multitudes in order to identify the consumers’ profiles in terms of their ecologically conscious behavior. We obtained 4 different groups of consumers in terms of the ECCB and perceived consumer effectiveness. The centers of classes were established at the level of the second iteration. The ANOVA table (see appendix 3) shows that the segmentation criteria are discriminant and produce this typology. The allotment of observations in groups is presented in table 7 as follows: Table 7 The final centers of classes obtained Items Classes 1 2 3 4 ECCB5 3,50 4,00 2,81 1,91 ECCB7 2,00 3,50 1,48 1,62 ECCB8 4,33 2,59 1,19 1,56 ECCB10 2,00 2,66 2,24 1,43 ECCB12 2,00 3,94 1,81 1,60 ECCB14 2,00 3,47 1,24 1,67 ECCB15 5,00 3,38 1,38 1,86 ECCB16 5,00 2,91 3,38 1,85 ECCB21 4,00 3,66 1,24 1,93 ECCB3 5,00 4,13 2,00 2,35 ECCB9 3,83 4,47 1,81 2,88 ECCB11 5,00 3,41 1,10 3,26 International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 368 Items Classes ECCB13 2,67 4,19 1,81 3,57 EPC1 3,67 4,34 2,05 3,86 EPC2 2,00 2,53 1,57 1,87 EPC3 3,83 3,66 2,90 3,48 EPC4 4,00 4,34 2,19 3,89 Size 6 32 21 141 Percentage 3% 16% 10,5% 70,5% Table 7 shows that the obtained class 1 constitutes 3% of our sample. It is characterized by both « an ecologically conscious purchasing behavior » and « an energy saving and recycling behavior » that are significantly high. This group of individuals is almost convinced that it can reduce the problems of pollution and pretends taking into consideration the ecological impact of their purchases. These are the « aware » customers. The second obtained class constitutes 16% of our sample. It is characterized by a high « ECCB » and by a high « energy saving and recycling behavior». This group of individuals is particularly persuaded that it can reduce pollution related problems and takes into consideration the ecological impact that his purchases could have. We here speak of the « convinced » customers. The 3rd obtained class constitutes 10.5% of our sample. This group of individuals is persuaded that it could reduce pollution problems. However, when there is a need to buy an ecological product, this group shows a weak conscious behavior as opposed to a relatively average energy saving and recycling behavior. It is the case of the « mitigated » customers. The fourth obtained class constitutes the most important section of our sample. It is composed of 70,5% of observations and is characterized by a weak ecologically conscious behavior with a weak energy saving and recycling behavior. This group of individuals is weakly persuaded that it can reduce pollution problems. These are the « resistant » customers. Furthermore, it is worth specifying that all the obtained classes are based on the personal variables of the sample. The conducted Chi-square tests (see appendix 4) help study the variation of the obtained classes based on gender, age, level of instruction, CSP, monthly income, religiosity, place of residence and civil status. The results are significant for all the variables at a risk of 5%. The examination of the table in appendix 4 shows that the socio-demographic profile of the various classes is as follows: Class 1 (the aware) is exclusively composed of young, single female student, aged 20 to 34, religious with a monthly income of less than 400 dinars and living in a rural environment. Class 2 (the convinced) is essentially composed of married or single women (either students or housewives) the majority of whom are aged between 20 and 34 years old with a higher education, mostly religious with a monthly income of less than 400 dinars and living in a rural environment. International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 375 Appendix 2 Sample characteristic Gender Percentage Male 47,5 Female 52,5 Age Percentage 20 -34 46,0 35-44 27,5 45-59 10,0 60 and over 16,5 Education level Percentage Primary 13,0 Secondary 14,0 University 73,0 CSP Percentage Unemployed 4,5 Retired 11,5 Liberal profession 9,0 Employee 9,0 Managers and in highly intellectual professions 30,0 Student 26,0 Housewife 10,0 Monthly income (Unit : DT) Percentage Less than 400 37,5 400-800 24,0 800 -1200 7,0 1200 -1600 11,5 More than 1600 20,0 Religiosity Percentage Religious 40,5 Non-religious 59,5 Place of residence Percentage Urban environment 78,5 Rural environment 21,5 Civil status Percentage Single 36,5 Married 58,0 Divorced 5,5 International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 376 Appendix 3 Results of the classification of dynamic multitude ANOVA Class Error F Significance Mean squares ddl Mean squares ddl ECB3 41,250 3 ,897 196 45,998 ,000 ECB5 41,954 3 ,876 196 47,887 ,000 ECB7 32,135 3 ,440 196 72,969 ,000 ECB8 24,797 3 ,464 196 53,394 ,000 ECB9 34,836 3 ,842 196 41,384 ,000 ECB10 15,460 3 ,446 196 34,631 ,000 ECB11 37,587 3 1,331 196 28,246 ,000 ECB12 47,703 3 ,402 196 118,543 ,000 ECB13 26,513 3 ,949 196 27,930 ,000 ECB14 31,701 3 ,606 196 52,314 ,000 ECB15 40,393 3 ,712 196 56,706 ,000 ECB16 36,111 3 ,692 196 52,217 ,000 ECB21 39,025 3 ,349 196 111,959 ,000 EPC1 24,604 3 ,983 196 25,030 ,000 EPC2 4,948 3 ,610 196 8,112 ,000 EPC3 2,871 3 1,271 196 2,259 ,083 EPC4 21,705 3 ,830 196 26,156 ,000 Appendix 4 Specification of classes obtained based on personal variables Gender: Chi-square tests Value ddl Asymptotic significance (bilateral) Pearson chi-square 16,048a 3 ,001 Likelihood ratio 18,588 3 ,000 Linear association by linear 15,230 1 ,000 Number of valid observations 200 a. 2 cells (25.0%) have theoretical size less than 5. The minimum theoretical size is 2.85. Cross table Gender * Class assignment (dynamic clusters) Size Class assignment (dynamic clusters) Total 1 2 3 4 Gende r Male 0 9 7 79 95 Female 6 23 14 62 105 Total 6 32 21 141 200 International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 377 Age: Chi-square tests Value ddl Asymptotic significance (bilateral) Pearson chi-square 82,514a 9 ,000 Likelihood ratio 86,754 9 ,000 Linear association by linear ,341 1 ,560 Number of valid observations 200 a. 7 cells (43.8%) have theoretical size less than 5. The minimum theoretical size is .60. Education level: Chi-square tests Value ddl Asymptotic significance (bilateral) Pearson chi-square 68,615a 6 ,000 Likelihood ratio 63,476 6 ,000 Linear association by linear 4,892 1 ,027 Number of valid observations 200 a. 7 cells (58.3%) have theoretical size less than 5. The minimum theoretical size is.78. Cross table Education level * Class assignment (dynamic clusters) Size Class assignment (dynamic clusters) Total 1 2 3 4 Education level Primary 0 0 4 22 26 Secondary 0 0 14 14 28 University 6 32 3 105 146 Total 6 32 21 141 200 Cross table Age * Class assignment (dynamic clusters) Size Classe d'affectation (nuées dynamiques) Total 1 2 3 4 Age 20 - 34 6 24 3 59 92 35 -44 0 0 0 55 55 45 - 59 0 0 10 10 20 60 and over 0 8 8 17 33 Total 6 32 21 141 200 International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 378 CSP: Chi-square tests Value ddl Asymptotic significance (bilateral) Pearson chi-square 116,679a 18 ,000 Likelihood ratio 129,521 18 ,000 Linear association by linear 13,742 1 ,000 Number of valid observations 200 a. 17 cells (60.7%) have theoretical size less than 5. The minimum theoretical size is .27. Cross table CSP * Class assignment (dynamic clusters) Size Class assignment (dynamic clusters) Total 1 2 3 4 CSP Unemployed 0 0 0 9 9 Retired 0 8 0 15 23 Liberal profession 0 0 6 12 18 Employee 0 0 4 14 18 Managers and in highly intellectual professions 0 0 0 60 60 Student 6 14 3 29 52 Housewife 0 10 8 2 20 Total 6 32 21 141 200 Monthly income: Chi-square tests Value ddl Asymptotic significance (bilateral) Pearson chi-square 36,419a 12 ,000 Likelihood ratio 49,808 12 ,000 Linear association by linear 13,032 1 ,000 Number of valid observations 200 a. 10 cells (50.0%) have theoretical size less than 5. The minimum theoretical size is .42. International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 379 Cross table Monthly income * Class assignment (dynamic clusters) Size Class assignment (dynamic clusters) Total 1 2 3 4 Monthly income Less than 400 6 18 11 40 75 400 - 800 0 6 8 34 48 800 - 1200 0 0 2 12 14 1200 - 1600 0 0 0 23 23 More than 1600 0 8 0 32 40 Total 6 32 21 141 200 Religiosity: Chi-square tests Value ddl Asymptotic significance (bilateral) Pearson chi-square 29,667a 3 ,000 Likelihood ratio 39,059 3 ,000 Linear association by linear 7,302 1 ,007 Number of valid observations 200 a. 2 cells (25.0%) have theoretical size less than 5. The minimum theoretical size is 2.43. Cross table Religiosity * Class assignment (dynamic clusters) Size Class assignement (dynamic clusters) Total 1 2 3 4 Religiosity Religious 6 20 0 55 81 Non-religious 0 12 21 86 119 Total 6 32 21 141 200 Place of residence: Chi-square tests Value ddl Asymptotic significance (bilateral) Pearson chi-square 14,997a 3 ,002 Likelihood ratio 22,792 3 ,000 Linear association by linear 14,410 1 ,000 Number of valid observations 200 a. 3 cells (37.5%) have theoretical size less than 5. The minimum theoretical size is 1.29. International Journal of Management, Accounting and Economics Vol. 2, No. 5, May, 2015 ISSN 2383-2126 (Online) © IJMAE, All Rights Reserved www.ijmae.com 380 Cross table Place of residence * Class assignment (dynamic clusters) Size Class assignment (dynamic clusters) Total 1 2 3 4 Place of residence Urban environment 6 32 18 101 157 Rural environment 0 0 3 40 43 Total 6 32 21 141 200 Civil status: Cross table Civil status * Class assignment (dynamic clusters) Size Class assignements (dynamic clusters) Total 1 2 3 4 Civil status Single 6 14 3 50 73 Married 0 18 18 80 116 Divorced 0 0 0 11 11 Total 6 32 21 141 200 Chi-square tests Value ddl Asymptotic significance (bilateral) Pearson chi-square 20,890a 6 ,002 Likelihood ratio 25,778 6 ,000 Linear association by linear 5,969 1 ,015 Number of valid observations 200 a. 5 cells (41.7%) have theoretical size less than 5. b. The minimum theoretical size is .33.